The primary aim of this investigation is to study the interaction of aminoglycosides with the diseased kidney independent of changes in serum drug concentration - time profiles which result from a decreased renal clearance of drug. These studies will test the hypothesis that the diseased kidney undergoing compensatory hypertrophy is inherently more sensitive to aminoglycoside nephrotoxicosis than is the normal kidney. Drug dosage adjustment regimens in clinical use today assume that the normal serum threshold to nephrotoxicity is unchanged in the presence of renal disease. Chronic renal failure will be induced in dogs by subtotal surgical nephrectomy. Phase one will involve comparing the steady state renal clearance of gentamicin in normal and diseased dogs to determine whether renal disease causes and increase in the fractional tubular reabsorption of gentamicin, an event predicted by this hypothesis. Phase two will assess the comparative nephrotoxicity of gentamicin administered in identicle concentration-time profiles to normal and diseased dogs for 12 days. Drug will be administered by a computer controlled infusion pump which will generate the same drug concentration-time profiles to all dogs on the basis of low dose pharmacokinetic studies performed in each animal prior to infusions. The profile selected will not produce clinical nephrotoxicity in health dogs. Renal function will be evaluated before and after drug administration by a panel of quantitative renal function tests. Post-mortem evaluation will include microscopic pathology (light and electron microscopy), assay of tissue for gentamicin concentration, and biochemical analysis of pertinent tissue enzyme activity. In addition, a new procedure for examining these parameters in tissue preparations consisting of primarily proximal tubules will be evaluated. Phase three will apply the results of the previous phases to evaluate the nephrotoxic potential of five clinical methods of administering gentamicin to dogs with chronic renal failure. The above studies should better define the mechanisms of aminoglycoside-induced nephrotoxicosis in patients with renal insufficiency and provide information on constructing minimally toxic dose regimens for these patients. This information is needed in order to reduce the clinical incidence of aminoglycoside nephotoxicosis, currently estimated to be from 10 to 25% of human patients receiving the drug.